Bias-Policy Iteration Based Adaptive Dynamic Programming for First-Order Fully Actuated Systems

  • Huaiyuan Jiang
  • , Ruiqing Zhang*
  • , Bin Zhou
  • *Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

In this paper, the bias-policy iteration (bias-PI) based adaptive dynamic programming for first-order fully actuated systems is considered. By exploiting the fully actuated property, the bias-PI method for first-order fully actuated systems is revisited, with a more streamlined formulation and a more concise convergence proof provided. The data-driven implementation for the proposed algorithm is introduced by neural networks accordingly. A numerical example verifies the effectiveness of the proposed results.

Original languageEnglish
Title of host publicationProceedings of the 4th Conference on Fully Actuated System Theory and Applications, FASTA 2025
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages2109-2114
Number of pages6
ISBN (Electronic)9798331526924
DOIs
StatePublished - 2025
Externally publishedYes
Event4th Conference on Fully Actuated System Theory and Applications, FASTA 2025 - Nanjing, China
Duration: 4 Jul 20256 Jul 2025

Publication series

NameProceedings of the 4th Conference on Fully Actuated System Theory and Applications, FASTA 2025

Conference

Conference4th Conference on Fully Actuated System Theory and Applications, FASTA 2025
Country/TerritoryChina
CityNanjing
Period4/07/256/07/25

Keywords

  • Admissible control
  • Data driven control
  • Fully actuated systems
  • Policy iteration
  • Reinforcement learning

Fingerprint

Dive into the research topics of 'Bias-Policy Iteration Based Adaptive Dynamic Programming for First-Order Fully Actuated Systems'. Together they form a unique fingerprint.

Cite this